相关实验视频

Updated: Mar 3, 2026

Curation of Computational Chemical Libraries Demonstrated with Alpha-Amino Acids

Curation of Computational Chemical Libraries Demonstrated with Alpha-Amino Acids

Published on: April 13, 2022

双向神经网络用于根据属性对分子结构进行排序

Renato Frazzato Viana¹, Juarez L F Da Silva², Luis G Dias³

¹Center of Mathematics, Computation and Cognition, Federal University of ABC, Av. dos Estados, 5001, Santo André, São Paulo 09210-580, Brazil.

|March 2, 2026

概括

这项研究引入了一种使用罗网络进行分子排名的深度学习模型,加速了材料科学和药物开发方面的发现. 这种双向学习方法超越了某些分子性质预测的传统方法.

科学领域:

计算化学和材料科学计算化学和材料科学
机器学习在科学发现中的作用

背景情况:

传统的分子发现依赖于缓慢,昂贵的量子化学计算或实验选.
机器学习提供直接从结构中加速分子性质预测.
排名分子结构可以足以进行选,绕过精确的属性价值确定.

研究的目的:

开发和评估一个深度学习模型来对分子结构进行排名.
为了比较一个语网络的性能与对向学习与标准点向回归.
用Uni-Mol分子表示来评估模型的稳定性.

主要方法:

开发了一个使用语网络架构的深度学习模型.
采用双向学习来训练分子结构排名模型.
在QM7x和QO2Mol数据集上评估性能,与点向回归进行比较.

主要成果:

双向学习的罗网络在预测绝对能量性质 (例如总和轨道能量) 方面表现优于标准点向回归.
点向回归对于衍生性质 (例如,HOMO-LUMO间隙) 和非能量性质 (例如,二极子时刻) 仍然有效.
双向学习到等级的方法在各种Uni-Mol模型大小 (V1和V2) 中始终优于点向回归.

结论:

使用罗网络进行双向学习是一种强大而有效的分子结构排名方法,在特定属性预测中表现优于点向回归.
这种方法加速了分子查和发现,在诸如能量储存,催化和药物开发等领域.
该框架表现出一致的优越性,即使使用先进的预训练的变压器骨干,如Uni-Mol.

更多相关视频

A Protocol for Computer-Based Protein Structure and Function Prediction

A Protocol for Computer-Based Protein Structure and Function Prediction

Published on: November 3, 2011

Structure and Coordination Determination of Peptide-metal Complexes Using 1D and 2D 1H NMR

Structure and Coordination Determination of Peptide-metal Complexes Using 1D and 2D 1H NMR

Published on: December 16, 2013

相关概念视频

Predicting Molecular Geometry

Predicting Molecular Geometry

VSEPR Theory for Determination of Electron Pair Geometries

Molecular Shapes

Molecular Shapes

Molecules have characteristic shapes that are crucial for their function. The arrangement of various electron groups around the central atom dictates their molecular geometry. Electron pairs in the valence shell of a central atom will adopt an arrangement that minimizes repulsions between the electron pairs by maximizing the distance between them. The valence electrons form either bonding pairs, located primarily between bonded atoms, or lone pairs.
Two regions of electron density in a diatomic...

Molecular Models

Molecular Models

Physical models representing molecular architectures of chemical compounds play essential roles in understanding chemistry. The use of molecular models makes it easier to visualize the structures and shapes of atoms and molecules.

¹H NMR: Pople Notation

¹H NMR: Pople Notation

The Pople nomenclature system classifies spin systems based on the difference between their chemical shifts. Coupled spins are denoted by capital letters with subscripts indicating the number of equivalent nuclei. When the coupled nuclei have well-separated chemical shifts, they are assigned letters that are far apart in the alphabet, such as A and X. When the difference in chemical shifts is small, coupled nuclei are named using adjacent letters of the alphabet (AB, MN, or XY).
A proton...

Newman Projections

Newman Projections

Different notations are used to represent the three-dimensional structure of molecules on two-dimensional surfaces. One of the most commonly used representations is the dash-wedge formula. The dashed wedges, solid wedges, and the plane lines indicate the groups situated behind the plane, coming out of the plane, and in the plane, respectively.
The organic molecules rotate across the single bonds leading to numerous temporary three-dimensional structures of varying energy known as...

Molecular Orbital Theory II

Molecular Orbital Theory II

Molecular Orbital Energy Diagrams